SciELO - Scientific Electronic Library Online

vol.17 issue6Influence of unilateral fatigue of lower limbs on the bilateral vertical jumpEffect of continuous and interval physical exercise on weight and biochemical profile of pregnant Wistar rats and consequences on fetal body weight author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Revista Brasileira de Medicina do Esporte

Print version ISSN 1517-8692

Rev Bras Med Esporte vol.17 no.6 São Paulo Nov./Dec. 2011 



Autonomic modulation during incremental exercise with upper limbs in individuals with spinal cord injury



Ralmony de Alcantara SantosI; Flávio de Oliveira PiresI,II; Rômulo BertuzziI,II; Fernando Roberto de-OliveiraI,II,III; Adriano Eduardo Lima-SilvaI,II

IResearch Group in Sports Sciences. Nutrition College, Federal University of Alagoas - Maceió, AL
IIStudy Group in Aerobic Performance. Physical Education and Sports School, University of São Paulo - São Paulo, SP
IIIStudy Group of Human Movement. Physical Education Department, Federal University of Lavras - Lavras, MG

Mailing address




Individuals with spinal cord injury (SCI) show alterations in the autonomic regulation at rest which can affect the heart rate variability (HRV) during exercise. The aim of this study was to compare the HRV threshold of a group with SCI with two control groups. Twenty-four men, seven with SCI (24.8 ± 2.0 years, 76.5 ± 6.9kg, 176.3 ± 1.9cm), eight sedentary subjects without SCI (CONSED) (24.1 ± 1.8 years, 75.1 ± 3.6kg, 176.3 ± 3.0cm) and nine athletes without SCI (CONATH) (22.6 ± 1.4 years, 73.8 ± 5.3kg, 175.6 ± 2.5cm) volunteered in this study. The participants performed an upper limb incremental test on a cycle ergometer (17.2W/two minutes) until exhaustion. The SD1 index of HRV was measured every stage, and the HRV threshold was identified at: 1) the first workload that elicited SD1 values lesser than 3ms; 2) the first stage that elicited a difference between SD1 values from two consecutive stages lesser than 1ms. SD1 values at rest or at 30% of Wmax were higher (p < 0.05) in CONSED (45.8 ± 6.8ms) than CONATH group (19.5 ± 4.4ms) but none of them were different (p > 0.05) from group SCI (25.8 ± 4.5ms). The 3ms HRV threshold expressed in absolute values or in relative to maximal workload and heart rate was not different between groups. However, the workload of 1ms HRV threshold in control group (68.8 ± 8.3W) was significantly higher (p < 0.05) than SCI group (21.5 ± 4.3W), but these groups were not different from CONSED group (41.3 ± 8.7W). These results suggest changes in autonomic control during exercise in subjects with SCI, which might lead to an early parasympathetic withdrawal during incremental exercise.

Keywords: autonomic nervous system, exercise, paraplegia.




The cardiac rhythm, measured by the time passed between two consecutive electrocardiogram R waves and by the cardiac monitor (R-R interval), does not present uniform cycles during the time. Conversely, a temporal variation between the successive R-R intervals, which is called heart rate variability (HRV), occurs(1).

HRV has been used as a non-invasive means of evaluation of the heart neural control(2-5). Increase in the HRV represents higher activity of the parasympathetic nervous system (PNS) and lower activity of the sympathetic nervous system (SNS)(6). Some studies suggest that higher activity of PNS, represented by higher HRV, would be associated with better physical aptitude and lower incidence of developing cardiac diseases(7,8). Inversely, higher SNS activity, represented by lower HRV, would be related to higher index of morbidity and mortality(9).

Acute physical exercise is one of the factors which causes important acute alterations in the cardiac autonomic regulation. These alterations may be measured by an HRV index named SD1(1012), which is obtained by the immediate standard deviation of the Poincaré plot. This index quantifies the vagal activity, removing the influence of non-stationary tendencies in the temporal bout, as during incremental exercise. During an incremental exercise, there is gradual reduction of this index with the increase of exertion intensity, up to approximately 50% of peak load, with tendency to stabilization in the subsequent loads(13). Thus, this stabilization tendency probably occurs due to the fact that HRV decrease during exercise is continuously associated with the same phenomenon, that is, the removal of the vagal influence on the sinoatrial node. Therefore, the moment of this transition is named HRV threshold (HRVT) and can be an important indicator of the transition from the PNS activity domain to the SNS,13-16).

The statement that the HRVT represents the transition from the PNS to the SNS lets us investigate how the clinical conditions affect the autonomic regulation during exercise. For example, individuals with spinal cord injury (SCI) present deficit in motor control and limited capacity of performing daily tasks, which increases the incidence of cardiac complications and alterations in the neural control of the cardiac muscle (17). Some studies report that individuals with spinal cord injury present HRV reduction(18,19)and increase of the heart rate (HR) at rest(20). These results suggest increase in the SNS control and PNS decrease in injured individuals at the rest condition.

Although the experimental evidence suggests autonomic alterations at rest derived from the SCI (18,21,22), little is known about these possible alterations during exercise. Thus, the information obtained in the literature lets us presume that, individuals with SCI would have lower HRV during the incremental exercise test and, therefore, HRVL occurring in lower workloads, in comparison to a group of non-injured individuals. However, until the present moment, none study has compared the HRV during exercise between individuals with and without SCI. Thus, the aim of the present study was to analyze the HRV response and the HRVL during a maximal upper limbs progressive test on a cycle ergometer (ULPT). The hypothesis of the present study is that individuals with SCI would have lower HRV when compared with individuals with out SCI, resulting in decrease of intensity corresponding to the HRVT.




24 male individuals were evaluated, who were divided in: 1) men with SCI, wheelchair basketball practitioners (n = 7); 2) noninjured men, sedentary for upper limb activities (CONSED, n = 8); and 3) non-injured men, combat sports which involve upper limbs athletes (CONATH, n = 9). The SCI group was composed of two subjects with partial injury and five with total injury. Out of these, four had injury above T6 and three below it. The two last groups (CONSED and CONATH) were used as control groups, being a control group considered untrained and the other trained for upper limbs activities (ULPT). The physical characteristics of the three groups are described in table 1. Each of the participants was informed about the risks associated with the study protocol and signed a consent form agreeing on participating in the experiment. This study was approved by the Ethics in Research Committee of the Federal University of Alagoas.



Experimental outlining

Data collection procedure wasperformed in three occasions. On the first, the individuals answered a questionnaire corresponding to their health status. On the second, the individuals went through an anthropometric evaluation for measurement of body mass (kg) and stature (cm), being used, respectively, mechanical scale and a stadiometer (Welmy-Industry and Sales LTDA, Model R-110). The body mass and stature measurement of the individuals with SCI was performed through the patients' self-report, who had a medical follow-up. On the third, the individuals performed a ULPT. The individuals received recommendation not to perform vigorous physical activity, not to ingest caffeinated (coffee, chocolate, mate, guaraná powder or Coke) or alcoholic substances in the 24 hours preceding the tests.

Incremental test

The individuals arrived at the laboratory and wore an elastic band with a wrist cardiac monitor (Polar S810i, Polar Electro OY®, Kempele, Finland). They remained laid for five minutes for recording of the HRV at rest. Subsequently, a two-minute warmup was performed using only the inertial resistance of the cycle ergometer adapted to upper limbs (Monark®, Stockholm Sweden) and an increment of 17W was given at every two minutes until exhaustion. Exhaustion was determined when the individual could not keep the pre-set rhythm, which was of approximately 70rpm. HR was monitored beat-after-beat during the entire incremental test. Subjective perceived exertion was measured through the Borg's scale (15 points) at the end of each stage(23).

Maximal HR (HRmax) was considered as the highest value reached during the test. The maximum load reached (Wmax) was considered as the last complete stage in the test before voluntary exhaustion. When the last reached stage was not completed, the Wmaxwas obtained from the following modified equation(24):



Where Wmaxis equal to the maximum load reached in the test, Wcomplete corresponds to the load of the last complete stage, and final time to the time of the last incomplete stage.



After the end of the test, the data were stored in the monitor, transferred to a computer and assessed with the use of specific software (Polar Precision Performance, Polar Electro OY®, Kempele, Finland). Each R-R interval was plotted in relation to the previous R-R interval (Poincaré plot) in a time window corresponding to the last rest minute andof each stage of the incremental test. These intervals generate a sequence of points in an ellipse shape. The transversal axis of the ellipse, termed standard deviation "1" (SD1), represents the immediate variability of the R-Rintervals(13,14,25,26).

The SD1 index of each stage was then plotted in relation to the intensity. The HRVT was identified from two criteria: 1) on the first load corresponding to a value lower than three milliseconds(5,15), termed HRVT3ms; and 2) on the first stage in which the difference between the SD1 of two consecutive stages was lower than one millisecond, termed HRVT1ms(25). The HR corresponding to the threshold load was obtained from the linear correlation between HR-intensity. The thresholds were expressed in absolute values (W and bpm) and related to the Wmax and HRmax ( % Wmax and % HRmax respectively).

In order to verify possible alterations in the autonomic regulation during each stage of the test between the different groups, the SD1 index was also compared at rest, in the absolute workloads referring to the warm-up, 17.2W and 34.4W, as well as in the loads concerning the maximum load (30%, 60%, 90% and 100%). These loads were chosen for comparison for representing the distinct phases of the incremental test (low, moderate and high intensities).



Data were expressed as mean ± mean standard error (MSE). Analysis of variance repeated measures one-way ANOVA was used to verify the differences of the dependent variables (HRmax, Wmax, HRVT3ms, HRVT1ms) between groups. Repeated measures 3 x 4 ANOVA (intensity x group) was also performed to compare the absolute and relative loads between groups. When differences were identified through ANOVA, the comparisons were followed by the Scheffé post hoctest. The analyses were performed using the statistical package SPSS (version 13.0), and significance level lower than 5% was adopted (p < 0.05).



The age, stature and body mass values did not present significant differences among the SCI, CONSED and CONATH groups (p > 0.05).

Significant differences in the HR max and Wmax between groups (p > 0.05) were not found in the comparisons of the variables obtained during the incremental test (table 2). The HRVT3ms, both in absolute (W and bpm) and relative values (%Wmax and %HRmax), were not significantly different between groups (p > 0.05). However, the HRVT1ms, when expressed in absolute load (W), was significantly higher in the CONATH group (p < 0.05) than in the SCI group, but no significant differences have been found between CONSED and CONATH groups, or between CONSED and CSI groups (p > 0.05).



When SD1 was compared in different intensities, significant differences were found between groups only at rest (figure 1). The CONSED group presented SD1 values significantly higher than the CONATH group (p < 0.05), but none of them was different from the SCI group (p > 0.05). Likewise, the SD1 at 30% of Wmax was significantly lower in the CONATH group than in the CONSED group (p < 0.05). However, significant differences have not been found between CONATH and SCI groups or between CONSED and SCI groups (p > 0.05) (figure 2).






The aim of this study was to compare the HRV and the HRVT between among the SCI, CONSED and CONATH groups. It was found that the CONSED group presented higher SD1 values than the CONATH group both at rest and at 30% of Wmax, but in none of these groups the values were different from the ones found in the SCI group. The HRVT3ms expressed in absolute (W and bpm) and relative values (%Wmax and %HRmax) was not different between groups; however, the HRVT1ms load in the CONATH group was significantly higher than in the SCI group.

In the present investigation, the HRVT3ms did not present significant differences between groups. However, the HRVT1ms load in the SCI group was lower than in the CONATH group, suggesting PNS removal and SNS early activation in this group. These findings are in agreement with the ones found by Jacobs et al.(20), who compared HR in individuals with and without SCI, demonstrating that the individuals with SCI had higher HR during exercise. A probable physiological explanation for this would be the low peripheral vasoconstriction of individuals with SCI and the absence of a lower limb active muscle pump(27), both being able to limit the venous return to the cardiac muscle, leading to reduction in the final ventricular diastolic volume as well as cardiac contraction efficiency restriction(28,29). In a trial to make up for these alterations, there may be an early SNS activation in the cardiac muscle of individuals with SCI, justifying the HRVT1ms in lower loads in the SCI group.

Another explanation for these findings is that the CONATH group would present higher training level than the SCI group. Data by Fronchettiet al.(5) demonstrated that the HRVT increases significantly after a three-week period of high-intensity interval training. Therefore, the SD1 vs. intensity curve was dislocated to the right and to the top, suggesting that this kind of training results in delay in PNS removal during the progressive exercise. Nevertheless, in the present study, both groups (SCI and CONATH) were composed of individuals trained for upper limbs, in a trial to control the training effects in the comparisons performed. Since the HRVT1ms of both groups was not different from the ones found in the CONSED group, the differences found between SCI and CONATH groups cannot be explained by the training level. Such statement corroborates the previous justification that SCI individuals present autonomic imbalance, and consequently, overload in the cardiac pump, increasing hence the SNS activity.

At rest and at 30% of Wmax conditions, the CONSED group presented SD1 significantly higher than in the CONATH group. Apparently, there is not a logical explanation for these results, since it is well demonstrated in the literature that trained individuals would have higher HRV than sedentary ones(16). However, some studies also suggest that when trained individuals surpass a suitable training limit, alterations in the autonomic balance are observed, which reflect on HR increase and HRV decrease(6). Yet, measurements which mark training excess have not been performed in this investigation, since this was not the focus of this investigation. Consequently, it is not possible to state that these differences have been actually derived from this phenomenon.

On the other hand, it was expected that the SCI group presented lower HRV values at rest than the other groups(18,19). However, significant differences between the SCI and the other groups have not been found for the HRV at rest in the present study. Probably, these controversial results compared to the literature(18,19)may be due to the physical fitness level of the individuals from the SCI group, who were wheelchair basketball practitioners. Some studies suggest that individuals with better physical fitness present higher HRV at rest(25). Corroborating this statement, Dixonet al. and Janssen et al.(8,30) verified that trained individuals, when being compared with untrained individuals, present higher HRV during rest. Thus, the fact the SCI group is composed of wheelchair athletes may explain the reason why the SD1 values at rest were similar to the ones found in the other groups.

In conclusion, the results of the present study suggest that the subjects with SCI present a HRVT1ms in lower loads and, therefore, it seems reasonable to state that there are alterations in the autonomic control during exertion in this group, suggesting early PNS removal during the incremental exercise.


To João Paulo Lopes da Silva and Tenildo Lopes for the help in the data collection phase and finally, to the participants of the study sample.



1. Reis AF, Bastos BG, Mesquita ET, Romêo LJM, Nóbrega CL. Disfunção parassimpática, variabilidade da freqüência cardíaca e estimulação colinérgica após infarto agudo do miocárdio. Arq Bras Cardiol 1998;70:193-9.         [ Links ]

2. Bootsma M, Swenne CA, Van Bolhuis HH, Chang PC, Cats VM, Bruschke AV. Heart rate and heart rate variability as indexes of sympathovagal balance. Am J Physiol 1994;266:1565-71.         [ Links ]

3. Stauss HM. Heart rate variability. AJP- Regul Integ Comp Physiol 2003;285:927-31.         [ Links ]

4. Hautala A. Effect of physical exercise on autonomic regulation of heart rate [dissertation]. [Oulu, Finland]. Faculty of Medicine; 2004:78.         [ Links ]

5. Fronchetti L, Nakamura FY, De-Oliveira FR, Lima-Silva AE. Effects of high-intensity interval training on heart rate variability during exercise. J Exerc Physiol Online 2007;10:1-9.         [ Links ]

6. Achten J, Jeukendrup AE. Heart rate monitoring: applications e limitations. Sport Med 2003;33:517-38.         [ Links ]

7. Almeida MB, Araújo CGS. Effects of aerobic training on heart rate. Rev Bras Med Esporte 2003;9:104-12.         [ Links ]

8. Dixon EM, Kamath MV, McCartney N, Fallen, EL. Neural regulation of heart rate variability in endurance athletes and sedentary controls. Cardiovasc Res 1992;26:713-9.         [ Links ]

9. Alonso DO, Forjaz CLM, Rezende LO, Braga AMFW, Barreto ACP, et al. Comportamento da freqüência cardíaca e da sua variabilidade durante as diferentes fases do exercício físico progressivo máximo. Arq Bras Cardiol 1998;71:787-92.         [ Links ]

10. Gallo JrL, Maciel BC, Marin-Neto JA, Martins LE, Lima-Filho EC, Golfetti R, et al. Control of heart rate during exercise in health and disease. Braz J Med Biol Res 1995;28:1179-84.         [ Links ]

11. Mitchell JH. Neural control of the circulation during exercise. Med Sci Sports Exerc 1986;22:141-54.         [ Links ]

12. Rowell LB. Human circulation: regulation during physical stress. New York: Oxford University Press; 1986:174-406.         [ Links ]

13. Lima JRP, Kiss MAPD. Limiar de Variabilidade da freqüência cardíaca. Rev Bras Ativ Fis Saúde 1999;4:29-38.         [ Links ]

14. Nakamura FY, Aguiar CA, Fronchetti L, Aguiar AF, Lima JRP. Alteração do limiar de variabilidade da frequência cardíaca após treinamento aeróbio de curto prazo. Motriz (UNESP), Rio Claro 2005;11:1-10.         [ Links ]

15. Brunetto AF, Roseguini BT, Silva BM, Hirai DM, Ronque EV. Limiar de variabilidade da freqüência cardíaca em adolescentes obesos e não-obesos. Rev Bras Med Esporte 2006;14:145-9.         [ Links ]

16. Cambri LT, Fronchetti L, De-Oliveira FR, Gevaerd MS. Variabilidade da freqüência cardíaca e controle metabólico. Arq Sanny Pesq Saúde 2008;1:72- 82.         [ Links ]

17. Jacobs PL, Nash MS. Exercise recommendations for individuals with spinal cord injury. Sports Med 2004;34:727-51.         [ Links ]

18. Bunten DC, Warner AL, Brunnemann SR, Segal JL. Heart rate variability is altered following spinal cord injury. Clin Auton Res 1998;8:329-34.         [ Links ]

19. Claydon VE, Krassioukov AV. Clinical correlates of frequency analyses of cardiovascular control after spinal cord injury. Am J Physiol Heart Circ Physiol 2008;294:668-78.         [ Links ]

20. Jacobs PL, Mahoney ET, Robbins A, Nash M. Hipokinetic circulation in persons with paraplegia. Med Sci Sports Exerc 2002;34:1401-7.         [ Links ]

21. Inoue K, Ogata H, Hayano J, Miyake S, Kamada T, Kuno M, et al. Assessment of autonomic function in traumatic quadriplegic and paraplegic patients by spectral analysis of heart rate variability. J Auton Nerv Syst 1995;54:225-34.         [ Links ]

22. Steinberg LL, Lauro FAA, Sposito MMM, Tufik S, Mello MT, Naffah-Mazzacoratti MG, et al. Catecholamine response to exercise in individuals with different levels of paraplegia. Braz J Med Biol Res 2000;33:913-8.         [ Links ]

23. Borg G. Psychophysical bases of perceived exertion. Med Sci Sports Exerc 1987;14:377-81.         [ Links ]

24. Kuipers H, Verstappen FTJ, Keizer HA, Guerten P, Van Kranenburg P. Variability of aerobic performance in the laboratory and its physiologic correlates. Int J Sports Med 1985;6:197-201.         [ Links ]

25. Tulppo MP, Mäkikallio TH, Seppänen T, Laukkanen RT, Huikuri HV. Vagal modulation of heart rate during exercise effects of age and physical fitness. Am J Physiol 1998;274:424-9.         [ Links ]

26. Mourot L, Bouhaddi M, Perrey S, Cappelle S, Henriet MT, Wolf JP, et al. Decrease in heart rate variability with overtraining: assessment by the Poincaré plot analysis. Clin Physiol Funct Imaging 2004;24:10-8.         [ Links ]

27. Wecht JM, Meersma, RE, Weir JP, Bauman WA, Grimm DR. Effects of autonomic disruption and inactivity on venous vascular function. Am J Physiol Heart Circ Physiol 2000;278:515-20.         [ Links ]

28. Figoni SF. Exercise response and quadriplegia. Med Sci Sports Exerc 1993;25:433-44.         [ Links ]

29. Jacobs PL, Nash MS, Klose KJ, Guest RS, Needham-Shropshire BM, Green BA. Evaluation of a training program for persons with SCI paraplegia using the Parastep ambulation system: Part 2. Effects on physiological responses to peak arm ergometry. Arch Phys Med Rehabil 1997;78:794-8.         [ Links ]

30. Janssen MJ, Bie J, Swenne CA, Oudhof J. Supine and standing sympathovagal balance in athletes and controls. Eur J Appl Physiol 1993;67:164-7.         [ Links ]



Mailing address:
Faculdade de Nutrição (FANUT) - Universidade Federal de Alagoas (UFAL)
Avenida Lorival Melo Mota, S/N Campus A.C. Simões
57072-970 Tabuleiro do Martins Maceió, AL, Brazil

All authors have declared there is not any potential conflict of interests concerning this article.

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License